Early Warning Signals for OpenVLA Failure under Visual Distribution Shift
Quick summary
arXiv:2606.29699v2 Announce Type: replace-cross Abstract: Visual shifts can cause a vision-language-action policy to fail after initially plausible behavior. We ask whether OpenVLA's internal activations contain signals associated with the steps before failure. We freeze the policy, record one MLP activation per LIBERO-10 step, and fit two linear monitors. Occlusion reduces task success from $57\%$ to $17\%$. Within failed matched-reset trajectories, a layer-16 logistic probe attains AUROC $0.972$ and AUPRC $0.352$, whereas action disagreement attains AUROC $0.496$. Without refitting, the occl
Key takeaways
- arXiv:2606.29699v2 Announce Type: replace-cross Abstract: Visual shifts can cause a vision-language-action policy to fail after initially plausible behavior.
- We ask whether OpenVLA's internal activations contain signals associated with the steps before failure.
- We freeze the policy, record one MLP activation per LIBERO-10 step, and fit two linear monitors.
Why it matters
The significance is not only the legal text but how it changes product design. Decisions around “Early Warning Signals for OpenVLA Failure under Visual Distribution Shift” may reshape data collection, model training, output accountability and market access.
Member comments